Diagnostic Blood Biomarkers for Acute Pulmonary Embolism: A Systematic Review

Abstract

(1) Background: The current diagnostic algorithm for acute pulmonary embolism (PE) is associated with the overuse of CT pulmonary angiography (CTPA). An additional highly specific blood test could potentially lower the proportion of patients with suspected PE that require CTPA. The aim was to summarize the literature on the diagnostic performance of biomarkers of patients admitted to an emergency department with suspected acute PE. (2) Methods: Medline and Embase databases were searched from 1995 to the present. The study selection process, data extraction, and risk of bias assessment were conducted by two reviewers. Eligibility criteria accepted all blood biomarkers except D-dimer, and CTPA was used as the reference standard. Qualitative data synthesis was performed. (3) Results: Of the 8448 identified records, only 6 were included. Eight blood biomarkers were identified, of which, three were investigated in two separate studies. Red distribution width and mean platelet volume were reported to have a specificity of ≥ 90% in one study, although these findings were not confirmed by other studies. The majority of the studies contained a high risk of selection bias. (4) Conclusions: The modest findings and the uncertain validity of the included studies suggest that none of the biomarkers identified in this systematic review have the potential to improve the current diagnostic algorithm for acute PE by reducing the overuse of CTPA.

Keywords: biomarker; diagnostic accuracy; pulmonary embolism.

Figure 1

Preferred reporting items for systematic reviews and a meta-analysis flowchart.

Figure 2

Risk of bias tabular (A) [26,27,28,29,30,31] and diagram (B) summary: review of authors’ judgments on each bias domain in the quality assessment of diagnostic accuracy studies-2 revised tool [24] for each included study.

Figure 3

Numerical and graphically (paired forest plots) presentation of the biomarkers’ sensitivity, specificity, and 95% confidence intervals [26,27,28,29,30,31].